In-line QC checkpoints

By Mohamad Sinno, Charcoal Expert · published 7 September 2026. Not reviewed by the Head of QC — the reviewer byline on our older production pages carries the date those pages were read, and this one has not been read yet.

This is the reference version of the checkpoint system: one row per checkpoint, with what is sampled, what is recorded, which method applies where one is defined, and what happens when a check fails. Where something is not recorded, the row says so rather than leaving the cell to be read as a criterion.

The short answer

Six of the nine checkpoints are formal sampling stages. At those, pieces are pulled at random, at least five pieces per 50 kg, and weight, ash content, moisture, ash colour are recorded per sample. Two of the checks have written, repeatable methods. One checkpoint uses a threshold we do not publish, and one check on the list has no recorded method at all — both are marked.

The reference table

The two methods written down in full

Both are stated with their conditions because a method without them cannot be repeated or disputed — a drop test with no height, count and surface is a sentence, and a moisture figure with no temperature and duration is a number. The in-house moisture check is a quick check and is labelled one everywhere it appears; the laboratory's constant-mass determination is a different procedure with a different answer.

Two cells that are honestly empty

Checkpoint 2's threshold. A numeric carbonization-completeness threshold is used at checkpoint 2 and is not published: what it measures, and in what units, has not been recorded — and a number without a stated basis is not a specification.

Checkpoint 8's odour check. It is on the roster and it is performed. Its stage detail and its method are not recorded anywhere we can publish from — no procedure, no pass criterion. Several suppliers publish a smell test; adopting one and presenting it as ours would fill this cell with somebody else's practice. It stays marked until a real method is written down, and the odour page says the same thing at more length.

And the burn-behaviour check at the same checkpoint is a gate, not a number. It passes or it does not. No duration is recorded for publication and none is published anywhere on this site — the conditions under which the factory measures burning time have never been written down, which is set out on the burn-time page.

What a failure does

One rule covers every row: when an in-process check fails — moisture, drop test, cracks, odour — the batch, or the failing part of it, is rejected and does not ship. No re-drying-and-retest route is claimed, because none is operated.

It is worth setting that against what this market publishes, because the difference is real and runs the other way from the usual marketing direction. Suppliers describe reprocessing a failed batch, adjusting production parameters and running it again, with outright rejection as a last resort; one publishes a drop-test criterion satisfied by a fraction of the sample rather than by all of it; another releases on a stated moisture limit. Those are their practices, dated and attributed, and none of them is ours. We are not describing a recovery path we do not have — which costs us the more reassuring sentence and is the reason to believe the rest of the page.

Auditing against it

1. Start at the carton, not the table. The batch code identifies the production batch whose QC file holds the laboratory COA and the factory’s own results. The code is a four-digit batch production code, two letters identifying the packing team, and one check digit.
2. Ask for the batch file by that code. It holds the laboratory COA and the factory’s own results for the batch: ash content, moisture, burning time, drop test, cracks, temperature.
3. Repeat the two methods above on pieces from the same carton, and compare against the file rather than against the grade specification.
4. Ask which checkpoint a disputed value came from. Moisture appears at four of the nine, and a figure means different things at grinding, at oven exit and at packing.
5. Raise anything inside the claims window — ours is seven days from receipt, with replacement or credit as the remedy.

The drying log and the container loading log are not published, and no sample page of either has been supplied. So an audit here is against the file and the certificate, not against a published log — and the full buyer-side procedure is the batch audit protocol.